Abstract
BACKGROUND/AIMS: The safety of direct oral anticoagulants (DOACs) compared with warfarin in patients with both nonvalvular atrial fibrillation (AF) and clinically confirmed liver cirrhosis (LC) has not been well studied. We compared the risk of a major bleeding event between DOAC and warfarin treatments in this patient population. METHODS: A total of 238 cirrhotic patients with AF were retrospectively analyzed. The major bleeding event risk was compared between DOAC- and warfarin-treated groups. The median follow-up duration was 5.6 years. RESULTS: Among the 238 study patients with LC and AF, 128 (53.8%) received DOACs and 110 (46.2%) received warfarin. The mean patient age was 68.8 years, and 78.2% were men. A major bleeding event occurred in 10 and 20 patients in the DOAC and warfarin groups, respectively, most commonly caused by gastrointestinal bleeding (70.0%). The cumulative risk of major bleeding did not differ between the groups by log-rank test (p = 0.12). This finding did not change when using 60 propensity score-matched pairs. A multivariable Cox regression model indicated that the concomitant use of antiplatelet agents (adjusted hazard ratio [aHR], 2.06; 95% confidence interval [CI], 1.00 to 4.30; p = 0.048) and presence of esophageal or gastric varices confirmed by endoscopic examination (aHR, 2.31; 95% CI, 1.03 to 5.17; p = 0.04) were associated with major bleeding in the entire cohort. CONCLUSION: A major bleeding event risk is not increased by DOAC compared with warfarin treatment. Antiplatelet agent use and varices are independently associated with a higher risk of major bleeding during anticoagulation.